Reduced-Complexity Near-ML Detector for a Coded DSTTD-OFDM System

Abstract

This letter introduces an efficient near-maximum likelihood (ML) detector for a coded double space-time transmit diversity-orthogonal frequency division multiplexing (DSTTD-OFDM) system. The proposed near-ML detector constructs a candidate vector set through a relaxed minimization method. It reduces computational loads from $\mathcal{O}(2|\mathcal{A}|^{2})$ to $\mathcal{O}(|\mathcal{A}|^{2})$, where $|\mathcal{A}|$ is the modulation order. Numerical results indicate that the proposed near-ML detector provides both almost ML performance and considerable complexity savings.